feat: enable zero-downtime deployments with Helm and PodDisruptionBudget

- Add PodDisruptionBudget template with minAvailable: 1
- Fix RollingUpdate strategy: maxUnavailable: 0, maxSurge: 1
- Enable PDB in production values
- Update prod workflow to use helm upgrade instead of oc rollout restart

Author: danieltruong

.github/workflows/deploy-to-prod.yaml

@@ -47,6 +47,11 @@ jobs:
     needs: validate
     runs-on: ubuntu-latest
     steps:
+      - name: Checkout repository at version tag
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.version }}
+
       - name: Install OpenShift CLI
         uses: redhat-actions/openshift-tools-installer@v1
         with:
@@ -74,10 +79,23 @@ jobs:
           oc -n ${{ env.OPENSHIFT_NAMESPACE_TOOLS }} tag \
             ${{ env.IMAGE_NAME }}:${{ inputs.version }} ${{ env.IMAGE_NAME }}:prod
 
-      - name: Deploy to prod
+      - name: Install Helm
+        run: |
+          curl -fsSL https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3 | bash
+          helm version
+
+      - name: Deploy with Helm
         run: |
-          oc -n ${{ env.OPENSHIFT_NAMESPACE_PROD }} rollout restart deployment/${{ env.APP_NAME }}
-          oc -n ${{ env.OPENSHIFT_NAMESPACE_PROD }} rollout status deployment/${{ env.APP_NAME }} --timeout=5m
+          helm upgrade --install ${{ env.APP_NAME }} ./helm/${{ env.APP_NAME }} \
+            --namespace ${{ env.OPENSHIFT_NAMESPACE_PROD }} \
+            --values ./helm/${{ env.APP_NAME }}/values-prod.yaml \
+            --set image.tag=prod \
+            --wait --timeout=5m
+
+      - name: Verify deployment
+        run: |
+          echo "Deployment successful!"
+          oc get pods -n ${{ env.OPENSHIFT_NAMESPACE_PROD }} -l app.kubernetes.io/name=${{ env.APP_NAME }}
           
           echo "Deployment successful!"
           oc get pods -n ${{ env.OPENSHIFT_NAMESPACE_PROD }} -l app.kubernetes.io/name=${{ env.APP_NAME }}

helm/eagle-admin/templates/deployment.yaml

@@ -18,8 +18,8 @@ spec:
   strategy:
     type: RollingUpdate
     rollingUpdate:
-      maxSurge: 25%
-      maxUnavailable: 25%
+      maxSurge: 1
+      maxUnavailable: 0
   template:
     metadata:
       {{- with .Values.podAnnotations }}

helm/eagle-admin/templates/pdb.yaml

@@ -0,0 +1,13 @@
+{{- if and .Values.podDisruptionBudget.enabled (gt (int .Values.replicaCount) 1) }}
+apiVersion: policy/v1
+kind: PodDisruptionBudget
+metadata:
+  name: {{ include "eagle-admin.fullname" . }}
+  labels:
+    {{- include "eagle-admin.labels" . | nindent 4 }}
+spec:
+  minAvailable: {{ .Values.podDisruptionBudget.minAvailable | default 1 }}
+  selector:
+    matchLabels:
+      {{- include "eagle-admin.selectorLabels" . | nindent 6 }}
+{{- end }}

helm/eagle-admin/values-prod.yaml

@@ -15,6 +15,11 @@ resources:
     cpu: 100m
     memory: 256Mi
 
+# Pod Disruption Budget - ensure at least 1 pod during disruptions
+podDisruptionBudget:
+  enabled: true
+  minAvailable: 1
+
 env:
   DEPLOYMENT_ENVIRONMENT: "prod"
   BANNER_COLOUR: ""

helm/eagle-admin/values.yaml

@@ -80,3 +80,8 @@ autoscaling:
   minReplicas: 2
   maxReplicas: 4
   targetCPUUtilizationPercentage: 80
+
+# Pod Disruption Budget
+podDisruptionBudget:
+  enabled: false
+  minAvailable: 1